Open win10ogod opened 1 year ago
之後有計畫新的訓練功能嗎? 例如: 1.類似AlphaGo Zero的自我对弈 2.多模態 3.huggingface數據集 等有潛能但項目沒有的功能
只有多模态在考虑之列。
AlphaGo Zero
您認為AlphaGo Zero的思路能否用於對話式模型的學習([類似Google 的Gemini 模型),或RWKV模型用於設計芯片的潛能如何?
AlphaGo Zero
您認為AlphaGo Zero的思路能否用於對話式模型的學習([類似Google 的Gemini 模型),或RWKV模型用於設計芯片的潛能如何?
也許RWKV模型用在機械設計也會是不錯的方向呢
AlphaGo Zero
您認為AlphaGo Zero的思路能否用於對話式模型的學習([類似Google 的Gemini 模型),或RWKV模型用於設計芯片的潛能如何?
也許RWKV模型用在機械設計也會是不錯的方向呢
看路标, 目前规划太大的功能不现实。 当前能将简单业务闭环已经需要花很久的时间实现。
之後有計畫新的訓練功能嗎? 例如: 1.類似AlphaGo Zero的自我对弈 2.多模態 3.huggingface數據集 等有潛能但項目沒有的功能